partner selection in supply chain of vietnam’s textile and

Research ArticlePartner Selection in Supply Chain of Vietnamrsquos Textile andApparel Industry The Application of a Hybrid DEA and GM(11) Approach

Chia-NanWang12 Han-Khanh Nguyen1 and Ruei-Yuan Liao3

1Department of Industrial Engineering and Management National Kaohsiung University of Applied SciencesNo 415 Chien Kung Road Sanmin District Kaohsiung City 80778 Taiwan2Department of Industrial Engineering and Management Fortune Institute of Technology Kaohsiung 83160 Taiwan3National Sun Yat-sen University Kaohsiung 80424 Taiwan

Correspondence should be addressed to Han-Khanh Nguyen nguyenhankhanhgmailcom

Received 2 July 2017 Revised 25 September 2017 Accepted 10 October 2017 Published 8 November 2017

Academic Editor Emilio Jimenez Macıas

Copyright copy 2017 Chia-Nan Wang et alThis is an open access article distributed under theCreativeCommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Vietnam is currently among the top-five textile and apparel exporters and the industry is considered quite attractive to foreigninvestors Nevertheless the global textile and garment industry is experiencing important changes The three main producingregions in the world are China Southwest Asia (India Pakistan Bangladesh and Turkey) and ASEAN In order to maintain itspositioning and to establish stable and sustainable Vietnam textile and apparel development there must be radical changes Dueto this necessity the authors conducted this study by using grey forecasting to predict and reflect the condition of businesses in theperiod of 2017ndash2020 together with combining a DEAmodel to help businesses select the most appropriate strategic partner in thesupply chain in order to achieve economic goals and promote the strength of the businesses partaking in the association Besidesthis helps businesses exploit market opportunities and take advantage of the capabilities of the textile and apparel industry

1 Introduction

11 Overview of Vietnamrsquos Textile and Apparel Industry Aftermore than 20 years of continuous development with anaverage growth rate at 17 per year the textile and apparelindustry has become a major economic industry and is aleading export industry in Vietnam accounting for 10ndash15gross domestic product (GDP) [1] By the end of 2014 therewere 5214 textile and apparel companies in Vietnamwith themajority of those being small- andmedium-sized companiesThe textile and apparel work force occupiedmore than 20ofthe nation-wide work force and nearly 5 of the nation-widework force Garment companies accounted for the largestshare (84) followed by textile and spinning companies(15) Vietnam was one of the top 10 textile exporters in theworld with the market share at 31 in 2014 The countryrsquosmain export markets included the United States EuropeanUnion (EU) Japan and Korea (85 of the total exportgoods) providing the garment for mostly the low andmiddle

classes (see Table 1) Although foreign direct investment(FDI) enterprises represented only 25 in the volume theycontributed to more than 65 of Vietnamrsquos export turnover[2]

Currently Vietnam is involved in the process of thelowest value-added that is the ldquocutndashmakendashtrimrdquo with thesimple outworks production method (the ldquocutndashmakendashtrimrdquocomprises 85 see Table 2) Vietnamrsquos garment sector stilldepends on imported raw materials (70ndash80) mainlyfrom China Taiwan and Korea The size of the Vietnamesetextile and apparel market is limited at only $3 billionand per capita expenditures for the textile and apparel arerather low [2] Also the Vietnamese garment enterprises areencountering difficulties in occupying the domestic marketas those companies have to compete with fake and non-quota imported goods Branding includes just a few smalland successful firms the majority of Vietnamese textile andapparel enterprises are inadequate in absorbing knowledgesuch as building distribution systems design and branding

HindawiMathematical Problems in EngineeringVolume 2017 Article ID 7826840 16 pageshttpsdoiorg10115520177826840

2 Mathematical Problems in Engineering

Table 1 Main export markets of Vietnamrsquos textile and apparel in2016

Nation Unit ValueUnited States of America Billions of US Dollars 1145Japan Billions of US Dollars 29Korea Billions of US Dollars 2284China Millions of US Dollars 825Germany Millions of US Dollars 7262England Millions of US Dollars 715Holland Millions of US Dollars 538Sources [14]

Free trade agreements such as the Vietnam-Korea FreeTrade Agreement (VKFTA) European UnionndashVietnam FreeTrade Agreement (EVFTA) and Trans-Pacific PartnershipAgreement (TPP) open up many opportunities for theexport turnover growth in Vietnamrsquos textile and apparelindustry The industry is expected to reach $50ndash$55 billionin 2025 thanks to FTA [2] On the other hand Vietnam hasto face many difficulties and challenges in obeying originrules labor standards strict social responsibilities eco labelsenvironmental protection and so on Currently Vietnam isrelying on the supply of main fabrics and fibers from Chinaand Taiwan those that are not in the TPP Thus when theagreements come into force Vietnam may have little chanceto inherit tax incentivesThe textile and apparel industry withthe majority of small- and medium-sized companies havinglow financial capacity and competitiveness will experiencedifficulties when the economy is opened Hence in order topromote the overall strength of the industry the implementa-tion of coalition among textile enterprises has to be driven inforce for the enterprises to complete and move up to a highervalue chain aiming to gain benefits from alliances for exam-ple achieving economies of scale reducing production costsand transaction costs increasing access to technology newproduction resources and markets leveraging negotiationposition and competitiveness improving the organizationaland knowledge capacity through the sharing of experiencewithin coalition group and building risk reduction Joiningalliances is to better serve the market It is the only way fornew textile and apparel enterprises to provide a large numberof products with consistent quality and timely delivery formany partners and integrate deeper and wider into the worldeconomy

12 Global Textile and Apparel Value Chain Gereffi intro-duced the global textile and apparel value chain in 2002[3] Whereby the textile and apparel value chain is affectedby purchasers creating finished products must go throughmany stages and production activities are often carried outin many countries In particular well-known manufacturersbig wholesalers and retailers play key roles in establish-ing production networks and shaping mass consumptionthrough a series of strong brands and outsourcing activitiesto satisfy this demand The global textile and apparel valuechain is divided into five basic stages (1) supply of rawmaterials including natural cotton thread and so on (2)

production of intermediate goods products of this stageare fibers and fabrics provided by weaving knitting anddyeing companies (3) design and manufacture of finishedproducts implemented by garment companies (4) export bycommercial intermediaries (5) marketing and distribution[3] (See Figure 1)

In order to study and present the context of the wholesupply chain more research is required Within this studywe feature an in-depth study of consumer products for textilecompanies and suppliers for apparel companies Moreoverthis study ismainly focused on partner selection which is justa part of the supply chain

13 Literature Review In 1989 Deng introduced the greysystem theorywhich is an interdisciplinary scientific area andhas become quite common in systems to process and predictpartially unknown parameters [4]The grey model (GM) canestimate the behavior under grey system theory in the futurebased on a limited amount of available data The GM (11) isone of the most important parts of grey system theory andis considered to be the core of the grey prediction modelThe advantage of this model is that it can be used when theamount of data is insufficient to perform statistical analysismethods It requires only a small amount of data and randomsample data to calculate and produce the forecast resultsMany studies have applied the grey system For example in1989 Bunn conducted a study about forecasting with morethan one model [5] In 2001 Chen et al studied applicationof an innovative combined forecasting method in power [6]In 2005 Yamaguchi et al studied new grey relational analysisfor finding the invariable structure and its applications [7]In 2009 Lingbin et al used the grey theory model topredict the number of international tourists in China [8]Besides some researchers have conducted strategic alliancestudies For example Buyukozkan et al (2008) provideda decision support to make a careful assessment of an e-logistics partner [9] This research proposed a multicriteriadecision-making approach to effectively evaluate e-logisticsbased on strategic alliance partners Candace et al (2011)affirmed strategic alliances as an efficient means to access theresources needed for innovation in dynamic environments[10] Several researches used hybrid DEA (data developmentanalysis) and GM (11) to study strategic alliance whichrevealed high reliability This was a complementary conceptwhich assumed that two complementary companies to begood partners could make an alliance and perform betterbusiness Because this method can estimate the behavior ofdata in the future based on a limited amount of availabledata this complementary concept appeared in the hybridDEA andGM (11) to study strategic alliance for the electronicmanufacturing service industry [11] In 2016 Wang et alused DEA and a grey theory model to analyze the selectionstrategic alliance partner in the automobile industry [12] In2017 Wang and Nguyen used the grey and DEA model tostudy enhancing urban development quality based on theresults of appraising efficient performance of investors inVietnam [13]

Strategic alliances between organizations are now ubiq-uitous it has been used in the past for enterprises that have

Mathematical Problems in Engineering 3

Table 2 Overview of Vietnamrsquos textile and apparel industry

Indicators Unit ValueNumber of companies Companies 6000Enterprise scale People SMEs of 200ndash500 + Account for a large proportionCompany structure based on ownership Private (84) FDI (15) State-owned (1)

Company structure based on operation Sewing (70) spinning (6) weavingknitting (17) dying (4)ancillary industries (3)

Geographical allocation of company North (30) Central and plateau (8) South (62)Number of employees People 25 MillionAverage income per worker VND 45 MillionNumber of working days per week Day 6Number of hours worked per week Hour 48Number of shifts per day Shift 2Value of textile export in 2016 US$ 23841 BillionMethod of production Cut - Make ndash Trim 85 others 15Sources [2]

Productionnetworks

Exportnetworks

All retailoutlets

US garment factories(designing cuttingsewing buttonholingironing)

Cotton woolsilk etc

Fabric(weavingknittingfinishing)

Yarn(Spinning)

Domestic andMexicanCaribbeanbasin subcontractors

Brand-namedapparelcompanies

Overseas buyingoffices

Department stores

Massmerchandisechains

Specialty stores

Oil natural gas

Syntheticfibers

Petrochemicals

Asian garmentcontractors

Domestic and overseassubcontractors

Trading

companies

Natural fibers

Synthetic fibers

North America

Asian

All retail outlets

Textile companies Retail outletsApparel manufacturers

Marketingnetworks

Componentnetworks

Discount chains

Off-price factoryoutlet mail orderothers

Textile companies

Figure 1 Global textile and apparel value chain [3]

the same expertise and targets In the current economy ofintegration and development along with the limited capacityof enterprises the technology has not been developed nor hasit met the quality and quantity requirements of partners whocontinue to operate individually thus they will encounterdifficulties when facing competitors If we combine eachotherrsquos strengths we will create a value chain in each productline offer prestigious and quality products and enhancecompetitiveness in the domestic and international markets

Therefore strategic alliances between enterprises are urgentThus combining a hybrid DEA model and grey systemtheory helps us to choose the strategic alliance for textile andapparel enterprises to make the best of their business goalsand possibly exploit the opportunities of the environmentaddress challenges and take advantage of the capacity of thetextile and apparel industry in Vietnam This study proposesan effective new method Through the realities in Vietnamrsquostextile and apparel industry we would recommend finding


Table 3 List of Companies in textile and apparel industry

Order DMUs Companies name

1 DMU1 Hue Textile Garment Joint Stock Company(JSC)

2 DMU2 SaiGon Garment Manufacturing Trade JSC3 DMU3 TNG Investment and Trading JSC4 DMU4 Nha Be Garment Corporation JSC5 DMU5 Ha Dong Textile JSC6 DMU6 DongNai Garment Corporation7 DMU7 HoaTho TextilendashGarment JSC8 DMU8 PhanThiet Garment ImportndashExport JSC9 DMU9 Saigon 3 Garment JSC10 DMU10 ThangLoi International Garment JSC11 DMU11 HaNoi Industrial Textile JSC12 DMU12 HaNoi Textile and Garment JSC13 DMU13 March 29 Textile Garment JSC14 DMU14 GHOME Textile Investment JSC15 DMU15 Viet Tien Garment JSCSource synthetic by researcher [15ndash29]

the alliance partners for companies to solve those existingproblems that they cannot overcome when doing businessby using grey theory GM (11) to predict what would happenabout their business situation in the period 2017ndash2020 Afterthat we use a DEAmodel to evaluate ranking and score of alldecision-making units (DMU) in 2016 Combining analyticalresults from super-SBM model we put out analysis alliancesuggests bringing high economic efficiency for enterprises aswell as Vietnamrsquos textile and apparel industry

2 Materials and Methods

21 DMUs Collection After researching Vietnamrsquos textile andapparel industry this study reveals 15 companies collectedfrom the General Statistics Officersquos website [15ndash29] We didnot choose other enterprises because of this reason thecompany scale is small the uptime does not align with thedata requirements and the business data of those compa-nies do not match with the demand of grey theory modelthat we used in this research The synopsis is shown inTable 3

22 InputsOutputs Collection The inputsoutputs elementwill directly affect the results of the analysis and evaluationFrom the literature review this research summarizes financefactors that were not previously considered This study care-fully selects four inputs and two outputs which are collectedfrom financial statements of 15 DMUs

(i) Input factors include the following

Total assets (TA) which reflect the entire valueof existing assets of the business in reportingCost of sold capital (CS) which is the totalinput cost of the enterprise to produce products

including the cost of inputs fuel costs machin-ery direct labor and other expensesSelling expenses (SE) which include all costsincurred in the process of consuming productsgoods or servicesGeneral and administration expenses (AE)which are the total costs of business manage-ment administrative management and othergeneral services related to the operation of thebusiness

(ii) Output factors include the following

Revenue of sales (RS) which reflects salesand services revenue of the enterprise in anaccounting period of production and businessactivities from transactions sales and servicingoperationProfit after tax (PT) which reflects the totalnet profit after tax relating to the enterprisersquosbusiness activities

The selected factors reflect wholly a firmrsquos assets costs andprofits which can be used as the foundation for analysis cal-culation and evaluation of this study Researcher syntheticsare shown in Tables 4 5 6 and 7

23 Research Development The authors use the GM (11)model to predict the business situation of all DMUs and thesuper-SBM-I-V model to select strategic alliance partners forthe textile business in the future The process of conductingthis study is shown in Figure 2

Step 1 (objectives identification) Based on the requirementsof professional managerial and practical needs of the textileand apparel industry the authors select the objectives of thestudy

Step 2 (literature review) By researching and exploringthe chosen topic the authors examine available resourcesidentifies research purposes and develops hypothesis forhisher research topic to implement the research If the topictime and method overlap with another study the authorsrevert to Step 1Step 3 (methodology)

Prediction Method There are many prediction methods andthe authors use the GM (11) to predict the data from theDMUs from 2017 to 2020 because this is a highly reliableprediction method and suitable with the data of DMUs

DEA Models There are many methods to choose an alliancepartner and the authors use the super-SBM-I-V a method ofchoosing the right strategic alliance partner and is popularnowadays

Step 4 (research planning) The authors plan to carry out themain tasks to manage time well and to scientifically controlthe progress


Table 4 Data of 15 DMUs in 2013 (currency unit 1000000 VND)

DMUs Inputs Outputs(I)TA (I)CS (I)SE (I)AE (O)RS (O)PT

DMU1 509991 1152459 42110 54446 1306653 30880DMU2 657777 1062372 18633 90818 1228479 57032DMU3 961199 962177 26726 102633 1186685 14057DMU4 1878363 2291270 199623 192508 2818958 88070DMU5 158855 225771 6666 9543 255462 11518DMU6 454291 898732 18704 39094 983038 29553DMU7 974923 2217252 70805 86102 2454787 48340DMU8 103426 142211 1202 5371 172208 21481DMU9 968574 1713273 38784 77527 1861925 36189DMU10 72358 118064 18227 14084 162747 6888DMU11 140479 398557 14232 24012 459383 11949DMU12 1633770 1232482 46967 72444 1395956 29079DMU13 263782 326142 10801 37282 396971 9796DMU14 201567 178629 2769 1554 196211 1127DMU15 2456738 4159024 226059 209707 4831173 237117Source synthetic by researcher [15ndash29]




Step 5 (DMUs collection) The authors collected informationof 15 Vietnam textile and apparel companies which was inline with research objectives through the General StatisticsOfficersquos website

Step 6 (inputsoutputs collection) Input factors include totalassets cost of capital sold cost of sale and general andadministration expenses

Output factors include revenue of sales and profit aftertax

Step 7 (grey prediction) The prediction sequence is as fol-lows

Forecasting Use business data of 15 DMUs from 2013 to 2016and applyGM(11) to predict business performance from2017to 2020

Check Accuracy The authors use mean absolute percentageerror (MAPE) to calculate predicted errors thus ensuring thepredicted data are accurate compared with the requirementsIn case the accuracy does not meet with the requirementsDMUs must be selected accordingly

Step 8 (check Pearson coefficient) The authors use thePearson coefficient to check the correlation between inputsand outputs Suppose the inputs and outputs have a positive








correlation the result is highly reliable However if thecorrelation coefficient is negative or equals zero the inputsand outputs must be reselected

Step 9 (DEA analysis (super-SBM-I-V)) The sequence ofDEA analysis according to the super-SBM-I-V model is asfollows

Before Alliance The actual results of the order and businessperformance of the DMUs serve as a basis for the authors tochoose future alliance partner

After Alliance From the preunion assessment the authorsselect the DMU target as a partner alliance with other

businesses When combining DMU target with 14 otherDMUs the authors use the SBM-I-V model to analyzeand obtain 29 plans to compare and evaluate analysis andcomments

Step 10 (partner selection conclusion) Based on the resultsof analysis after the alliance the authors proposed the mostappropriate alliance solution in the supply chain of theenterprises which brings the highest economic efficiency aswell as the affiliate groups that are not highly effective inhelping the businesses with suitable direction

24 Grey ForecastingModel Before using theGM(11)modelinitial data must be tested against the formula [30]


Objectives identification

Literature review

Method collection(Prediction method DEA model)

Research planning

DMUs collection

Forecasting

Check Pearson coefficient

Partner selectionconclusion

Check accuracy

Grey prediction

Before alliance

After alliance

DEA analysis

(Super-SBM-I-V)

Ok

No

Inputsoutputs collection

No

No

Ok

Ok

Figure 2 Research process

120575119894 = 119909(0) (119894 minus 1)119909(0) (119894) (119894 = 2 3 119899) (1)

If all the 120575119894 values are within 120575(0)(119894) = (119890minus2(119899+1) 1198902(119899+1)) wecan use the GM (11) model to forecast

The GM (11) model is calculated based on the followingdifferential equation [31]

119889119909(1) (119896)119889119896 + 119886119909(1) (119896) = 119887 (2)

Within the equation 119886 and 119887 are coefficients The initial datais considered as a value chain as follows

119883(0) = (119909(0) (1) 119909(0) (2) 119909(0) (119899)) 119899 ge 4 (3)

In this research 119883(0) is the actual business metrics of DMUsrecorded from 2013 to 2016 Data after being tested will becalculated using the following steps

Step 1 Calculate the 119883(1) by using the cumulative method[31]

120594(1) = (120594(1) (1) 120594(1) (2) 120594(1) (119899)) 119899 ge 4 (4)

where 120594(1)(1) = 120594(0)(1) 120594(1)(119896) = sum119896119894=1 120594(0)(119894) 119896 = 1 2 3 119899Calculate series 119885(1) (mean value) of adjacent neighbor120594(1) by [31]119885(1) = (119885(1) (1) 119885(1) (2) 119885(1) (119899)) 119899 ge 4 (5)

where 119885(1)(120581) = 05( 120594(1)(120581) + 120594(1)(120581 minus 1)) 119896 = 2 3 119899Step 2 Set the equation for the GM (11) model and calculatethe 119885(1) values [31]

119889119883(1) (119896)119889119896 + 119886119883(1) (119896) = 119887 (6)

where 119885(1)(120581) = 05(119909(1)(120581) + 119909(1)(120581 minus 1)) 119896 = 2 3 119899


Table 8 The grades of MAPE

MAPE valuation () le10 10 divide 20 20 divide 50 ge50Accuracy Excellent Good Qualified UnqualifiedSources [30]

Step 3 Calculate the parameter 119886 and parameter 119887Parameter 119886 and parameter 119887 of the GM (11) model are

calculated based on the least-squares method as follows [31]

119886 = [119886119887]119879 = (119861119879119861)minus1 119861119879119884119873 (7)

where 119861 = [ minus120572119885(1)(2) 1

minus120572119885(1)(119899) 1

] 119884119873 = [ 119883(0)(2)119883(0)(119899)

]Step 4 Set the formula to calculate the predictive value of themodel [31]

119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] 119890minus119886120581 + 119887119886(120581 = 1 2 3 )

(8)

Then calculate the predicted values of the GM (11) modelbased on the following formula [31]

119883(0) (119896 + 1) = 119909(1) (119896 + 1) minus 119909(1) (119896) (9)

where 119909(0)(1) = 119909(0)(1) (120581 = 1 2 3 119899)25 Evaluation of Volatility Forecasts In this paper we useMAPE which is a tool to measure the accuracy value instatistics to identify the grey prediction models with goodperformanceTheMAPE is small It is called the slacks-basedmeasure of efficiency (SBM) [32]

MAPE = 1119899 [119899sum119894=1

10038161003816100381610038161003816100381610038161003816119860 119894 minus 119865119894119860 119894

10038161003816100381610038161003816100381610038161003816 times 100] (10)

The MAPE is divided into our ranks as shown in Table 8

26 Nonradial Super Efficiency Model (Super-SBM) In thissection we propose a model by which we discriminatebetween such DMUs it is called the slacks-based measureof efficiency (SBM) Here a super-SBM is expanded by theaddition of the super efficiency to the DEA model with thepolarities of the inputs and outputs The 119899 DMUs with theinputoutput matrices (119883 119884) are used in this model [33]in which 119883 = (119909119894119895) isin 119877119898times119899 and 119884 = (119884119894119895) isin 119877119904times119899 120582is a nonnegative vector in 119877119899 The vectors 119878minus isin 119877119898 and119878+ isin 119877119904 represent an excess input and a short falling output

respectively [34] The SBM model is given by followingequation [35]

min 120588 = 1 minus (1119898)sum119898119894=1 119904minus119894 11990911989401 + (1119904)sum119904119894=1 119904minus119894 1199101198940st 1205940 = 119883120582 + 119878minus

1205740 = 119884120582 minus 119878+(120582 ge 0 119883 ge 0 119884 ge 0)

(11)

Suppose (119901lowast 120582lowast 119904minuslowast 119904+lowast) is the optimal condition ofSBM and (1205940 1205740) is SBM efficient of DMU When 119901lowast = 1so 119904minuslowast = 0 and 119904+lowast = 0 (or there is no excess input anda short falling output) Hence a super-efficiency model wasintroduced for ranking DMUs and it was defined by thefollowing formula [36]

min 120575 = (1119898)sum119898119894=1 1199091198941199091198940(1119904) sum119904119903=1 1199101199031199101199030st 119909 ge 119899sum

119895=1 =0

120582119895119909119895

119910 le 119899sum119895=1 =0

120582119895119909119895120594 ge 1205940120574 le 1205740120574 ge 0120582 ge 0

(12)

Suppose the denominator is 1 so that the objectivefunction is an orientation input of the super-SBMmodelTheobtained feedback value of the objective function is larger orequivalent to 1 It is straightforward then that the inputs arepositive however the outputs are able to become a negativevalue The proper solution to solve this problem is to useDEA-solver pro 41 manual [37] as follows

Suppose 119910119903119900 le 0 It has defined 120574+119903 and 120574+minus119903 by119910+119903 = max119895=1119899

119910119903119895 | 119910119903119895 gt 0 119910+minus119903 = min

119895=1119899119910119903119895 | 119910119903119895 gt 0 (13)

If there is no positive component in the output 119903 itbecomes 119910+119903 = 119910+minus119903 = 1 The element 119904+119903 1205741199030 is instead thefollowing while the 1205741199030 is unchanged


Table 9 Data of DMU1 from 2013 to 2016 (currency unit 1000000 VND)

Year Inputs Outputs(I)TA (I)CS (I)SE (I)AE (O)RS (O)PT

2013 509991 1152459 42110 54446 1306653 308802014 588788 1221869 46946 53530 1379742 351192015 606216 1309806 51544 53208 1480821 440632016 679185 1341165 52198 26851 1478313 42778Source synthetic by researcher [9]

When 120574+119903gt 120574+minus119903 the element is

119904+119903120574+minus119903 (120574+119903 minus 120574+119903 ) (120574+119903 minus 1205741199030)When 120574+119903= 120574+minus119903 the element becomes

119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)

Here 119861 represents a large positive value in the DEA-solver the value of 119861 is about 100

Given the fact that the denominator is certainly lowerthan 120574+minus119903 it increases with the decrease of the distance 120574+119903 minus1205741199030and vice versa Hence this model significantly affects thenonpositive output valueThe obtained score is constant andit is a unidimensional unit for using units of themeasurement[35]

3 Results

31 Results and Analysis of the Grey Forecasting of the OutputValues for All DMUs The GM (11) model is utilized topredict the input and output factors values for each decision-making unit from 2017 to 2020 We used factor total assets(TA) ndash input 1 of DMU1 in Table 9 to explain calculationprocedure in this part

First we use the GM (11) model for trying to forecast thevariance of primitive series

Create the primitive series

119883(0) = (509991 588788 606216 679185) (15)

Perform the accumulated generating operation (AGO)

119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)

Create the different equation of GM (11)

To find 119883(1) series the following mean obtained by themean equation is

119911(1) (2) = 05 times (509991 + 1098779) = 804385119911(1) (3) = 05 times (1098779 + 1704995) = 1401887119911(1) (4) = 05 times (1704995 + 2384180) = 20445875

(17)

Solve equationsTo find 119886 and 119887 the primitive series values are substituted

into the grey differential equation to obtain

588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)

Convert the linear equations into the form of a matrixLet

119861 = [[[minus804385 1minus1401887 1minus20445875 1

]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)

And then use the least-squares method to find 119886 and 119887[119886119887] = 120579 = (119861119879119861)

minus1 119861119879119910119873 = [ minus007345207246709] (20)

Use the two coefficients 119886 and 119887 to generate the whiteningequation of the differential equation

119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)




DMU1 72181476 141371489 5562768 2529603 154631861 4854408DMU2 105399739 140374917 6726471 16910878 172072530 6213040DMU3 229445562 188126448 3238004 16541122 226870220 10049851DMU4 313066407 414324939 37195657 50295015 512147213 5798682DMU5 28120084 19240624 672979 715480 22596351 324228DMU6 47535342 81676535 1938136 5002824 89587588 769420DMU7 230252694 321282185 10062619 14822535 357909418 7673092DMU8 20048746 29330049 188572 531681 32172330 2502034DMU9 142497401 184461997 4082227 9823586 200749091 16663140DMU10 9436797 10252246 1061972 1368616 14500642 1124091DMU11 22543967 45438031 3842352 5562926 59831891 4515167DMU12 245373630 197804844 5339632 9491026 225020003 5125886DMU13 104633133 90118993 830548 5581679 100874238 2995318DMU14 1798577 38858767 355742 642383 43597352 1798577DMU15 439786435 779879734 28760815 29648383 878635184 41830995Sources calculated by researcher

Find the prediction model from equation

119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] times 119890minus119886120581 + 119887119886= [509991 minus 5207246709(minus00734) ]times 119890minus(minus00734)120581 + 5207246709(minus00734)

= 7604289363 times 11989000734times119896minus 7094298363

(22)

Substitute different value of 119896 into the equation119883(1) (1) = 509991 119896 = 0119883(1) (2) = 108914430 119896 = 1119883(1) (3) = 171240671 119896 = 2119883(1) (4) = 238313767 119896 = 3119883(1) (5) = 310495243 119896 = 4119883(1) (6) = 388174161 119896 = 5119883(1) (7) = 471769214 119896 = 6119883(1) (8) = 561730983 119896 = 7

(23)

derive the predicted value of the original series according tothe accumulated generating operation and obtain

119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330

119909(0) (3) = 119909(1) (3) minus 119909(1) (2) = 62326242119909(0) (4) = 119909(1) (4) minus 119909(1) (3) = 67073095119909(0) (5) = 119909(1) (5) minus 119909(1) (4)

= 72181476mdashResult forecast of 2017119909(0) (6) = 119909(1) (6) minus 119909(1) (5)



= 89961769mdashResult forecast of 2020(24)

As with above process we have the result of all DMUsfrom 2017 to 2020 respectively in Tables 10 11 12 and 13

In this study the authors useMAPE to check the accuracyof forecasts to ensure appropriate predictive methods and asa basis for highly reliable assessments The results are shownin Table 14

This research uses a model-based forecasting approach(an approach to obtain information for the future with mod-eling tools) through resimulating the past and comparingvalues the forecast is calculated by actual data and themodel if the error is within the allowable limits then themodel is reliable and usable As shown in Table 14 averageMAPE of 15 DMUs is less than 10 Based on rules asshown in Table 8 the predicted results in this study havea high level of accuracy (average MAPE of 15 DMUs is








408) This confirms that the GM (11) model used in thisstudy is suitable with highly reliable predicted results andanalysis

32 Pearson Correlation In this study the authors use thesuper-SBM-I-V model to analyze and choose a strategicalliance partner for DMUs where the condition for usingDEA is the correlation coefficient which could not benegative or equal to 0 Hence before analyzing DEA theauthors used the Pearson correlation coefficient to determinethe data used in this study which is in accordance with theDEA standards Correlation coefficients are always in therange of (minus1) to (1) if this factor is as close to (1) it is a perfect

linear relation [37] The results are shown in Tables 15 16 17and 18

Results of the Pearson correlation coefficient from Tables15 to 18 show that the factors used in this study have a stronglinear relationship which is consistent with the conditionsof DEA and can be used for analysis In the super-SBM-I-V model strategic partners will be selected for DMUs in thefuture

33 Analysis Alliance

331 Analysis beforeAlliance To identify and select the targetDMU for alliance with other DMUs in the future the authors





Table 14 Average MAPE of 15 DMUs

DMUs Average MAPE ()DMU1 351DMU2 179DMU3 536DMU4 772DMU5 529DMU6 301DMU7 378DMU8 485DMU9 359DMU10 413DMU11 390DMU12 240DMU13 328DMU14 738DMU15 127Average MAPE of 15 DMUs 408 ()Source calculated by researcher

use actual business data of the DMUs and the analyticalapplications from the super SBM-I-V of DEA to evaluate thebusiness situation ofDMUs before the allianceThe results areshown in Table 19

332 Analysis after Alliance Based on the actual perfor-mance of the business in 2016 and the results of business anal-ysis obtained from the super-SBM-I-V software the authorsestablishe DMU5 as the alliance target for the business Ifwe choose low-rated and performance DMUs it will bedifficult to persuade other DMUs to join the alliance Whencombing DMU5 to other 14 DMUs the authors have 29

combinationsThe authors used software of a DEA-solver pro80-super-SBM-I-Vmodel and obtained the results as shownin Table 20

In this study DMU5 was chosen as the target DMU foralliances with other enterprises in the textile supply chainbecause DMU5rsquos business performance in 2016 was ineffec-tive (specifically DMU5Rank = 14 (1415 DMUs) DMU5Score =09998)Hence theDMU5management teamneeds to boldlychange the strategy and choose an alliance solutionwith otherDMUs to improve their business efficiency On the otherhand DMU5 (HaDong Textile JSC) established in 1956 hasa long history in the textile industry with a large market andmodern technology and can easily persuade other DMUs tojoin their alliance

Based on the ranking and efficiency score at Table 20 wecan separate into two groups good alliances cooperative andinappropriate alliances as follows

Group 1 After strategic alliances the DMUs get good busi-ness efficiency make the relationship among the coalitionparties better and expand market share Candidates needto own good characteristics and necessary consistency withonersquos desire in businessThere are 13 DMUs (DMU14 DMU8DMU9 DMU3 DMU1 DMU13 DMU7 DMU10 DMU6DMU11 DMU4 DMU2 DMU15) in this group which helpDMU5 to improve results after the strategic alliance AlliancesDMU5 + DMU14 DMU5 + DMU8 DMU5 + DMU9DMU5 + DMU3 DMU5 + DMU1 DMU5 + DMU13 DMU5+ DMU7 DMU5 + DMU10 DMU5 + DMU6 DMU5 +DMU11 DMU5+DMU4DMU5+DMU2DMU5+DMU15receive a score of more than 1 These are shown in Table 21

Group 2 In this group DMU12 combinedwithDMU5 showsno any progress after alliance Thus this combined will notbe priority with selection of a strategic alliance partner in thefuture as shown in Table 22


Table 15 Correlation of inputs and outputs in 2013

TA CS SE AE RS PTTA 10000 09077 08869 09286 09176 08134CS 09077 10000 08977 09064 09985 09108SE 08869 08977 10000 09218 09168 08809AE 09286 09064 09218 10000 09245 08329RS 09176 09985 09168 09245 10000 09175PT 08134 09108 08809 08329 09175 10000Source calculated by researcher







34 Discussion of Alliance Partner Selection DMUs includeDMU14 DMU8 DMU9 DMU3 DMU1 DMU13 DMU7DMU10 DMU6 DMU11 DMU4 DMU2 and DMU15 Butthese DMUs would not be willing to combine with DMU5because some companies have a score after an alliance andranking is reduced before an alliance

In alliances that bring high performance in group 1DMU9 (Saigon 3 garment joint stock company GATEXIM)before alliance has a ranking at 3 (315 DMUs) howeverafter the alliance its ranking is at 8 (829 DMUs) Hence

the effect given is immensely good specifically DMU5 inHanam (the large economic center in the northern region ofVietnam) specializes in the textile industry and the DMU9 inHoChiMinhCity (the largest economic center in the south ofVietnam) specializes in the garment industry Thus in termsof expertise these two companies joined the alliance betweentextile and apparel enterprises when the parties signed themain agreement which helps to build the supply chain byeliminating intermediaries between suppliers and consumersand shorten the length of the channel consumption Joining


Table 19 Rank and score before alliances

Rank DMU Score1 DMU14 494952 DMU8 216023 DMU9 144684 DMU10 137765 DMU1 116606 DMU3 115427 DMU7 103038 DMU11 102629 DMU13 1025110 DMU6 1024111 DMU4 1018312 DMU2 1010513 DMU15 1000014 DMU5 0999815 DMU12 08677Source calculated by researcher

Table 20 Performance ranking of virtual DMUs

Rank DMU Score1 DMU14 457222 DMU8 195743 DMU10 137764 DMU5 + DMU14 105265 DMU1 104986 DMU9 104737 DMU5 + DMU8 102938 DMU5 + DMU9 102929 DMU5 + DMU3 1026610 DMU11 1025411 DMU13 1025112 DMU3 1019313 DMU5 + DMU1 1018914 DMU6 1017015 DMU5 + DMU13 1009216 DMU5 + DMU7 1008417 DMU15 1006718 DMU2 1004319 DMU7 1004220 DMU5 + DMU10 1004021 DMU5 + DMU6 1003222 DMU4 1002623 DMU5 + DMU11 1002424 DMU5 + DMU4 1001925 DMU5 + DMU2 1000626 DMU5 + DMU15 1000027 DMU5 0999828 DMU5 + DMU12 0880229 DMU12 08533Source calculated by researcher

the alliance helps businesses ensure that the factors of supplyand consumption remain stable In terms of geographiclocation enterprises can deploy and expand their distributionnetwork so that they can take advantage of the availablenetwork and human resources of their partnersThis not onlyhelps to save cost but also reduces time entering a marketto the fullest Besides joining alliances also can increase thecustomer base by leveraging each otherrsquos customer systemsBecause every business has its own characteristics matchingits inherent potential allianceswill have their own advantagesto exploit and complement each other Hence in order tosurvive compete and develop an alliance is inevitably atrend because it gives firms greater value-added alliance thanstanding alone Hence in order to survive compete anddevelop an alliance is inevitably a trend because it givesfirms greater value-added alliance than standing alone togain greater economic benefits on a larger scale to increaseprestige and brand name to reduce costs to maximizebusiness advantages of parties involved and to developthe customer base and distribution network This will helpstrengthen the position and competitiveness in the marketand improve business efficiency

4 Conclusions

With what has been happening in the textile and apparelindustry the need for linking is becoming increasinglyurgent In this study the authors provide amethod for findingand selecting the right strategic partner for textile and apparelenterprises The selected plan is to promote the internalstrength of all participating enterprises while promoting thestrength of the alliance in accordance with high volumeproducts high quality international quality timely deliveryand competitive price needs Authorities can rely on theseresearch results to make the correct and appropriate strategicdecisions in helping Vietnamrsquos textile and apparel industry todevelop when integrating with the global economy

Although the paper shows that GM (11) is a flexibleand easy model of use to predict what would happen inthe future business and DEA is an efficient tool to helpbusinesses find the right strategic partner we still cannotdeny some restrictions about these two methods for furtherstudies Different DEA models and optimization algorithmcan also be tested to revealmore changes and other industriescan be studied by this model in the future [38] This studyonly focuses on a quantitative model The authors will domore research of external environmental factors in the futureThe comparisons with other quantitative and qualitativeapproaches will be a good research direction as well

Conflicts of Interest

The authors declare no conflicts of interest

Acknowledgments

This research was supported in part by MOST 105-2221-E-151-039 from the Ministry of Sciences and Technology The


Table 21 The good alliances

Virtual Target DMU5 Virtual alliance DifferenceAlliance Ranking (a) Ranking (b) (a) minus (b)DMU5 + DMU14 27 4 23DMU5 + DMU8 27 7 20DMU5 + DMU9 27 8 19DMU5 + DMU3 27 9 18DMU5 + DMU1 27 13 14DMU5 + DMU13 27 15 12DMU5 + DMU7 27 16 11DMU5 + DMU10 27 20 7DMU5 + DMU6 27 21 6DMU5 + DMU11 27 23 4DMU5 + DMU4 27 24 3DMU5 + DMU2 27 25 2DMU5 + DMU15 27 26 1Source calculated by researcher

Table 22 The inappropriate alliance

Virtual Target DMU5 Virtual alliance DifferenceAlliance Ranking (a) Ranking (b) (a) minus (b)DMU5 + DMU12 27 28 (minus1)Source calculated by researcher

authors also appreciate the support from Fortune Institute ofTechnology and National Kaohsiung University of AppliedSciences in Taiwan

References

[1] Taichinh Textile and apparel stocks crowned httptap-chitaichinhvn

[2] Virac JSC Markets of Vietnamrsquos Textile and Apparel httpviracresearchcom

[3] G Gereffi ldquoThe international competitiveness of Asian econ-omies in the apparel commodity chainrdquo ERD Working PaperSeries no 5 pp 1ndash34 2002

[4] J Deng ldquoControl problems of grey systemsrdquo Systems andControl Letters vol 1 no 5 pp 288ndash294 1982

[5] D Bunn ldquoForecasting with more than one modelrdquo Journal ofForecasting vol 8 no 3 pp 161ndash166 1989

[6] G Chen K Li T Chung H Sun and G Tang ldquoApplication ofan innovative combined forecasting method in power systemload forecastingrdquo Electric Power Systems Research vol 59 no 2pp 131ndash137 2001

[7] D Yamaguchi G D Li and M Nagai ldquoGrey relational analysisfor finding the invariable structure and its applicationsrdquo TheJournal of Grey System vol 8 no 2 pp 167ndash178 2005

[8] Y Lingbin Z Xia and W Jin ldquoPrediction of the number ofinternational tourists in China based on gray model (11)rdquo inProceedings of the 2009 International Conference on ElectronicCommerce and Business Intelligence (ECBI rsquo09) pp 357ndash360Beijing China June 2009

[9] G Buyukozkan O Feyzioglu and E Nebol ldquoSelection of thestrategic alliance partner in logistics value chainrdquo International

Journal of Production Economics vol 113 no 1 pp 148ndash1582008

[10] E Y Candace and A T Thomas ldquoStrategic alliances withcompeting firms and shareholder valuerdquo Journal ofManagementand Marketing Research vol 6 pp 1ndash10 2011

[11] C N Wang N T Nguyen T T Tran and B B Huong ldquoAstudy of the strategic alliance for EMS industry the applicationof a hybrid DEA and GM (1 1) approachrdquo The Scientific WorldJournal vol 2015 Article ID 948793 2015

[12] C-N Wang X-T Nguyen and Y-H Wang ldquoAutomobileindustry strategic alliance partner selection The application ofa hybrid dea and grey theory modelrdquo Sustainability vol 8 no2 article no 173 2016

[13] C-NWang andH-KNguyen ldquoEnhancing urban developmentquality based on the results of appraising efficient performanceof investorsmdasha case study in vietnamrdquo Sustainability vol 9 no8 p 1397 2017

[14] B A Hung Overview of export activities Vietnamrsquos Textile andApparel in 2016 httpcafefvn

[15] Hue Textile Garment Joint Stock Company Financial reporthttphuegatexcomvn

[16] SaiGon Garment Manufacturing Trade JSC Financial reporthttpwwwgarmexsaigon-gmccom

[17] TNG Investment and Trading Joint Stock Company Financialreport httpwwwtngvn

[18] Nha Be Garment Corporation Joint Stock company Financialreport httpswwwnhabecomvn

[19] Ha Dong textile joint stock company Financial reporthttpdethadongvn

[20] DongNai Garment Corporation Financial report httpwwwdonagamexcomvn


[21] HoaThoTextile Garment Joint Stock company Financial reporthttpwwwhoathocomvn

[22] Phan Thiet Garment Import Export JSC Financial reporthttpwwwphanthietgarmentcomvn

[23] Saigon 3 Garment Joint Stock Company Financial reporthttpwwwsaigon3comvn

[24] ThangLoi International Garment JSC Financial reporthttpwwwmaythangloicomvn

[25] HaNoi Industrial Textile Joint StockCompany Financial reporthttphaicatexcom

[26] HaNoi Textile and Garment Joint Stock Corporation Financialreport httpwwwhanosimexcomvn

[27] March 29 Textile Garment Joint Stock Company Financialreport httpwwwhachibacomvn

[28] G HOME Textile Investment Joint Stock Company Financialreport httpwwwcozincomvn

[29] Viet Tien garment joint stock corporation Financial reporthttpwwwviettiencomvn

[30] G D Li S Masuda and M Nagai ldquoAn optimal predictionmodel using taylor approximationmethodrdquoThe Journal of GreySystem vol 11 pp 91ndash100 2011

[31] J L Deng ldquoIntroduction to grey system theoryrdquoThe Journal ofGrey System vol 1 no 1 pp 1ndash24 1989

[32] M Yu C Wang and N Ho ldquoA grey forecasting approachfor the sustainability performance of logistics companiesrdquoSustainability vol 8 no 9 p 866 2016

[33] C D Lewis ldquoIndustrial and Business Forecasting MethodrdquoJournal of Forecasting vol 2 no 2 pp 194ndash196 1982

[34] K Tone ldquoA slacks-based measure of efficiency in data envelop-ment analysisrdquo European Journal of Operational Research vol130 no 3 pp 498ndash509 2001

[35] E Duzakin and H Duzakin ldquoMeasuring the performanceof manufacturing firms with super slacks based model ofdata envelopment analysis an application of 500 major indus-trial enterprises in Turkeyrdquo European Journal of OperationalResearch vol 182 no 3 pp 1412ndash1432 2007

[36] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[37] F-Y Lo C-F Chien and J T Lin ldquoA DEA study to evaluatethe relative efficiency and investigate the district reorganizationof the Taiwan Power Companyrdquo IEEE Transactions on PowerSystems vol 16 no 1 pp 170ndash178 2001

[38] H Yapıcı and N Cetinkaya ldquoAn improved particle swarmoptimization algorithm using eagle strategy for power lossminimizationrdquoMathematical Problems in Engineering vol 2017Article ID 1063045 11 pages 2017

Submit your manuscripts athttpswwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of


Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of


Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of


Mathematical PhysicsAdvances in

Complex AnalysisJournal of


OptimizationJournal of


CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of


Operations ResearchAdvances in

Journal of


Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Algebra

Discrete Dynamics in Nature and Society



Decision SciencesAdvances in

Journal of


Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of


Table 1 Main export markets of Vietnamrsquos textile and apparel in2016

Nation Unit ValueUnited States of America Billions of US Dollars 1145Japan Billions of US Dollars 29Korea Billions of US Dollars 2284China Millions of US Dollars 825Germany Millions of US Dollars 7262England Millions of US Dollars 715Holland Millions of US Dollars 538Sources [14]

Free trade agreements such as the Vietnam-Korea FreeTrade Agreement (VKFTA) European UnionndashVietnam FreeTrade Agreement (EVFTA) and Trans-Pacific PartnershipAgreement (TPP) open up many opportunities for theexport turnover growth in Vietnamrsquos textile and apparelindustry The industry is expected to reach $50ndash$55 billionin 2025 thanks to FTA [2] On the other hand Vietnam hasto face many difficulties and challenges in obeying originrules labor standards strict social responsibilities eco labelsenvironmental protection and so on Currently Vietnam isrelying on the supply of main fabrics and fibers from Chinaand Taiwan those that are not in the TPP Thus when theagreements come into force Vietnam may have little chanceto inherit tax incentivesThe textile and apparel industry withthe majority of small- and medium-sized companies havinglow financial capacity and competitiveness will experiencedifficulties when the economy is opened Hence in order topromote the overall strength of the industry the implementa-tion of coalition among textile enterprises has to be driven inforce for the enterprises to complete and move up to a highervalue chain aiming to gain benefits from alliances for exam-ple achieving economies of scale reducing production costsand transaction costs increasing access to technology newproduction resources and markets leveraging negotiationposition and competitiveness improving the organizationaland knowledge capacity through the sharing of experiencewithin coalition group and building risk reduction Joiningalliances is to better serve the market It is the only way fornew textile and apparel enterprises to provide a large numberof products with consistent quality and timely delivery formany partners and integrate deeper and wider into the worldeconomy

12 Global Textile and Apparel Value Chain Gereffi intro-duced the global textile and apparel value chain in 2002[3] Whereby the textile and apparel value chain is affectedby purchasers creating finished products must go throughmany stages and production activities are often carried outin many countries In particular well-known manufacturersbig wholesalers and retailers play key roles in establish-ing production networks and shaping mass consumptionthrough a series of strong brands and outsourcing activitiesto satisfy this demand The global textile and apparel valuechain is divided into five basic stages (1) supply of rawmaterials including natural cotton thread and so on (2)

production of intermediate goods products of this stageare fibers and fabrics provided by weaving knitting anddyeing companies (3) design and manufacture of finishedproducts implemented by garment companies (4) export bycommercial intermediaries (5) marketing and distribution[3] (See Figure 1)

In order to study and present the context of the wholesupply chain more research is required Within this studywe feature an in-depth study of consumer products for textilecompanies and suppliers for apparel companies Moreoverthis study ismainly focused on partner selection which is justa part of the supply chain

13 Literature Review In 1989 Deng introduced the greysystem theorywhich is an interdisciplinary scientific area andhas become quite common in systems to process and predictpartially unknown parameters [4]The grey model (GM) canestimate the behavior under grey system theory in the futurebased on a limited amount of available data The GM (11) isone of the most important parts of grey system theory andis considered to be the core of the grey prediction modelThe advantage of this model is that it can be used when theamount of data is insufficient to perform statistical analysismethods It requires only a small amount of data and randomsample data to calculate and produce the forecast resultsMany studies have applied the grey system For example in1989 Bunn conducted a study about forecasting with morethan one model [5] In 2001 Chen et al studied applicationof an innovative combined forecasting method in power [6]In 2005 Yamaguchi et al studied new grey relational analysisfor finding the invariable structure and its applications [7]In 2009 Lingbin et al used the grey theory model topredict the number of international tourists in China [8]Besides some researchers have conducted strategic alliancestudies For example Buyukozkan et al (2008) provideda decision support to make a careful assessment of an e-logistics partner [9] This research proposed a multicriteriadecision-making approach to effectively evaluate e-logisticsbased on strategic alliance partners Candace et al (2011)affirmed strategic alliances as an efficient means to access theresources needed for innovation in dynamic environments[10] Several researches used hybrid DEA (data developmentanalysis) and GM (11) to study strategic alliance whichrevealed high reliability This was a complementary conceptwhich assumed that two complementary companies to begood partners could make an alliance and perform betterbusiness Because this method can estimate the behavior ofdata in the future based on a limited amount of availabledata this complementary concept appeared in the hybridDEA andGM (11) to study strategic alliance for the electronicmanufacturing service industry [11] In 2016 Wang et alused DEA and a grey theory model to analyze the selectionstrategic alliance partner in the automobile industry [12] In2017 Wang and Nguyen used the grey and DEA model tostudy enhancing urban development quality based on theresults of appraising efficient performance of investors inVietnam [13]

Strategic alliances between organizations are now ubiq-uitous it has been used in the past for enterprises that have






Productionnetworks

Exportnetworks

All retailoutlets


Cotton woolsilk etc


Yarn(Spinning)




Department stores


Specialty stores

Oil natural gas

Syntheticfibers

Petrochemicals



Trading

companies

Natural fibers

Synthetic fibers

North America

Asian

All retail outlets


Marketingnetworks

Componentnetworks

Discount chains


Textile companies























































Literature review


Research planning

DMUs collection

Forecasting



Check accuracy

Grey prediction

Before alliance

After alliance

DEA analysis

(Super-SBM-I-V)

Ok

No


No

No

Ok

Ok


120575119894 = 119909(0) (119894 minus 1)119909(0) (119894) (119894 = 2 3 119899) (1)



119889119909(1) (119896)119889119896 + 119886119909(1) (119896) = 119887 (2)


119883(0) = (119909(0) (1) 119909(0) (2) 119909(0) (119899)) 119899 ge 4 (3)



120594(1) = (120594(1) (1) 120594(1) (2) 120594(1) (119899)) 119899 ge 4 (4)



119889119883(1) (119896)119889119896 + 119886119883(1) (119896) = 119887 (6)

where 119885(1)(120581) = 05(119909(1)(120581) + 119909(1)(120581 minus 1)) 119896 = 2 3 119899






119886 = [119886119887]119879 = (119861119879119861)minus1 119861119879119884119873 (7)

where 119861 = [ minus120572119885(1)(2) 1

minus120572119885(1)(119899) 1

] 119884119873 = [ 119883(0)(2)119883(0)(119899)


119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] 119890minus119886120581 + 119887119886(120581 = 1 2 3 )

(8)


119883(0) (119896 + 1) = 119909(1) (119896 + 1) minus 119909(1) (119896) (9)


MAPE = 1119899 [119899sum119894=1

10038161003816100381610038161003816100381610038161003816119860 119894 minus 119865119894119860 119894

10038161003816100381610038161003816100381610038161003816 times 100] (10)





1205740 = 119884120582 minus 119878+(120582 ge 0 119883 ge 0 119884 ge 0)

(11)



119895=1 =0

120582119895119909119895

119910 le 119899sum119895=1 =0

120582119895119909119895120594 ge 1205940120574 le 1205740120574 ge 0120582 ge 0

(12)



119910119903119895 | 119910119903119895 gt 0 119910+minus119903 = min

119895=1119899119910119903119895 | 119910119903119895 gt 0 (13)








119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)



3 Results




119883(0) = (509991 588788 606216 679185) (15)


119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)




(17)



588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)



]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)


minus1 119861119879119910119873 = [ minus007345207246709] (20)


119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)








(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of









Productionnetworks

Exportnetworks

All retailoutlets


Cotton woolsilk etc


Yarn(Spinning)




Department stores


Specialty stores

Oil natural gas

Syntheticfibers

Petrochemicals



Trading

companies

Natural fibers

Synthetic fibers

North America

Asian

All retail outlets


Marketingnetworks

Componentnetworks

Discount chains


Textile companies























































Literature review


Research planning

DMUs collection

Forecasting



Check accuracy

Grey prediction

Before alliance

After alliance

DEA analysis

(Super-SBM-I-V)

Ok

No


No

No

Ok

Ok


120575119894 = 119909(0) (119894 minus 1)119909(0) (119894) (119894 = 2 3 119899) (1)



119889119909(1) (119896)119889119896 + 119886119909(1) (119896) = 119887 (2)


119883(0) = (119909(0) (1) 119909(0) (2) 119909(0) (119899)) 119899 ge 4 (3)



120594(1) = (120594(1) (1) 120594(1) (2) 120594(1) (119899)) 119899 ge 4 (4)



119889119883(1) (119896)119889119896 + 119886119883(1) (119896) = 119887 (6)

where 119885(1)(120581) = 05(119909(1)(120581) + 119909(1)(120581 minus 1)) 119896 = 2 3 119899






119886 = [119886119887]119879 = (119861119879119861)minus1 119861119879119884119873 (7)

where 119861 = [ minus120572119885(1)(2) 1

minus120572119885(1)(119899) 1

] 119884119873 = [ 119883(0)(2)119883(0)(119899)


119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] 119890minus119886120581 + 119887119886(120581 = 1 2 3 )

(8)


119883(0) (119896 + 1) = 119909(1) (119896 + 1) minus 119909(1) (119896) (9)


MAPE = 1119899 [119899sum119894=1

10038161003816100381610038161003816100381610038161003816119860 119894 minus 119865119894119860 119894

10038161003816100381610038161003816100381610038161003816 times 100] (10)





1205740 = 119884120582 minus 119878+(120582 ge 0 119883 ge 0 119884 ge 0)

(11)



119895=1 =0

120582119895119909119895

119910 le 119899sum119895=1 =0

120582119895119909119895120594 ge 1205940120574 le 1205740120574 ge 0120582 ge 0

(12)



119910119903119895 | 119910119903119895 gt 0 119910+minus119903 = min

119895=1119899119910119903119895 | 119910119903119895 gt 0 (13)








119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)



3 Results




119883(0) = (509991 588788 606216 679185) (15)


119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)




(17)



588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)



]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)


minus1 119861119879119910119873 = [ minus007345207246709] (20)


119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)








(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of























































Literature review


Research planning

DMUs collection

Forecasting



Check accuracy

Grey prediction

Before alliance

After alliance

DEA analysis

(Super-SBM-I-V)

Ok

No


No

No

Ok

Ok


120575119894 = 119909(0) (119894 minus 1)119909(0) (119894) (119894 = 2 3 119899) (1)



119889119909(1) (119896)119889119896 + 119886119909(1) (119896) = 119887 (2)


119883(0) = (119909(0) (1) 119909(0) (2) 119909(0) (119899)) 119899 ge 4 (3)



120594(1) = (120594(1) (1) 120594(1) (2) 120594(1) (119899)) 119899 ge 4 (4)



119889119883(1) (119896)119889119896 + 119886119883(1) (119896) = 119887 (6)

where 119885(1)(120581) = 05(119909(1)(120581) + 119909(1)(120581 minus 1)) 119896 = 2 3 119899






119886 = [119886119887]119879 = (119861119879119861)minus1 119861119879119884119873 (7)

where 119861 = [ minus120572119885(1)(2) 1

minus120572119885(1)(119899) 1

] 119884119873 = [ 119883(0)(2)119883(0)(119899)


119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] 119890minus119886120581 + 119887119886(120581 = 1 2 3 )

(8)


119883(0) (119896 + 1) = 119909(1) (119896 + 1) minus 119909(1) (119896) (9)


MAPE = 1119899 [119899sum119894=1

10038161003816100381610038161003816100381610038161003816119860 119894 minus 119865119894119860 119894

10038161003816100381610038161003816100381610038161003816 times 100] (10)





1205740 = 119884120582 minus 119878+(120582 ge 0 119883 ge 0 119884 ge 0)

(11)



119895=1 =0

120582119895119909119895

119910 le 119899sum119895=1 =0

120582119895119909119895120594 ge 1205940120574 le 1205740120574 ge 0120582 ge 0

(12)



119910119903119895 | 119910119903119895 gt 0 119910+minus119903 = min

119895=1119899119910119903119895 | 119910119903119895 gt 0 (13)








119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)



3 Results




119883(0) = (509991 588788 606216 679185) (15)


119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)




(17)



588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)



]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)


minus1 119861119879119910119873 = [ minus007345207246709] (20)


119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)








(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of









119886 = [119886119887]119879 = (119861119879119861)minus1 119861119879119884119873 (7)

where 119861 = [ minus120572119885(1)(2) 1

minus120572119885(1)(119899) 1

] 119884119873 = [ 119883(0)(2)119883(0)(119899)


119883(1) (120581 + 1) = [120594(0) (1) minus 119887119886] 119890minus119886120581 + 119887119886(120581 = 1 2 3 )

(8)


119883(0) (119896 + 1) = 119909(1) (119896 + 1) minus 119909(1) (119896) (9)


MAPE = 1119899 [119899sum119894=1

10038161003816100381610038161003816100381610038161003816119860 119894 minus 119865119894119860 119894

10038161003816100381610038161003816100381610038161003816 times 100] (10)





1205740 = 119884120582 minus 119878+(120582 ge 0 119883 ge 0 119884 ge 0)

(11)



119895=1 =0

120582119895119909119895

119910 le 119899sum119895=1 =0

120582119895119909119895120594 ge 1205940120574 le 1205740120574 ge 0120582 ge 0

(12)



119910119903119895 | 119910119903119895 gt 0 119910+minus119903 = min

119895=1119899119910119903119895 | 119910119903119895 gt 0 (13)








119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)



3 Results




119883(0) = (509991 588788 606216 679185) (15)


119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)




(17)



588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)



]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)


minus1 119861119879119910119873 = [ minus007345207246709] (20)


119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)








(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of










119904+119903(120574+minus119903)2 119861 (120574+119903 minus 1205741199030)

(14)



3 Results




119883(0) = (509991 588788 606216 679185) (15)


119883(1) = (509991 1098779 1704995 2384180)119909(1) (1) = 119909(0) (1) = 509991119909(1) (2) = 119909(0) (1) + 119909(0) (2) = 1098779119909(1) (3) = 119909(1) (2) + 119909(0) (3) = 1704995119909(1) (4) = 119909(1) (3) + 119909(0) (4) = 2384180

(16)




(17)



588788 + 119886 times 804385 = 119887606216 + 119886 times 1401887 = 119887

679185 + 119886 times 20445875 = 119887(18)



]]]

120579 = [119886119887]

119884119873 = [[[588788606216679185

]]]

(19)


minus1 119861119879119910119873 = [ minus007345207246709] (20)


119889119909(1)119889119896 + 00734 times 119909(1) = 5207246709 (21)








(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of











(22)


(23)


119909(0) (1) = 119909(1) (1) = 509991119909(0) (2) = 119909(1) (2) minus 119909(1) (1) = 57915330





















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of
















































4 Conclusions





Acknowledgments








References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of










References
















































Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of






























Volume 2014




Journal of











Journal of


Function Spaces






Algebra





Journal of




partner selection in supply chain of vietnam’s textile and

Documents